Language Planning for a New Order 491
repetition^ occupation) negation? etc, would be free from one obstacle
which confronts anyone who sets out to learn one of the existing Aryan
languages
This advantage does not meet the objection • are such derivative
affixes really necessary^ To do justice to it we must distinguish between
different classes of derivative affixes One class may be called semantic
or meaningful The affix either modifies the meaning of the root to
which it is attached or does the work of a compound formation Clear-
cut qualifying affixes such as those which express repetition, negation,
precedence, etc, merely usurp the function of necessary mobile items
already on the word list Thus to r^-state is to state again, port-natal
means after birth, to tfzzVjudge, means to judge wrongly > and the man
in bake-man could do as much work as the accretion -er in baker.
Compounds such as textile workers^ steel workers^ wood workers^ etc,
are admittedly longer than words of the fisher, writer > baker class, but
postman, milkman^ iceman^ dustman, dairyman show that compounds
made from independent words need not be more long-winded than
derivatives By using derivative affixes of the Esperanto or Novial type
we add a new burden to learning without much gam of space or any
additional clarity.
Affixes of the other class merely label the grammatical behaviour of
a word Thus the -dom in wisdom or the -ment in arrangement respec-
tively endow an attribute which would otherwise behave as an adjective,
or a process which would otherwise behave as a verb, with the gram-
matical prerogatives of a thing For instance, we can speak of wisdom
in contradistinction to wise^ as it, and we can put the article a or the>
which never stand immediately in front of arrange^ before arrangement
This shunting disguises the fact that wisdom remains within the adjec-
tival world and means nothing more than wise behaviour. Some inter-
languages carry this much further, having a special affiy for each of the
parts of speech.
At first sight there seems to be little in favour of this device. A
plausible excuse is that there is a rough and ready, if far from perfect,
correspondence between parts of speech in an Aryan language and
the three pigeon-holes into which we squeeze the physical world.
Although we meet many exceptions to any functional definition of the
parts of speech, it is approximately true to say that a noun-label usually
points to what is thing or person, an adjective-label to what is a pro-
perty, a verb label to what is action in a statement. Such affixes there-
fore give the beginner a clue to the lay-out of a sentojce which contains